You've likely heard of Natural Language Processing. Sophisticated computer algorithms help automatically understand human-generated text. A growing field within NLP is semantic analysis and its application to social media content is just starting to mature beyond labs and classrooms.
Semantic analysis is capable of providing valuable insight into the meaning behind social media content, providing a clear picture of a brand's position in the eyes of its customers.
Sentiment analysis has become the marketer’s newest weapon in an arsenal of tools for extracting meaning from social media. Why? Because gauging what customers really think about your brand or products isn’t easy when the focus group is now millions of people strong. Finding out what influencers in London thought of your product launch or which voters in Leeds are approving of your candidate isn’t an easy task, particularly in a real-time scenario of thousands of tweets per minute.
Who is more likely to buy your product and who is about to switch over to your competitor? sento answers all of these questions with highly accurate sentiment analysis that goes beyond just positive negative and neutral to help you get to the heart of the sentiment. Many tools say they do sentiment analysis but without NLP, human meaning and intent isn’t understood. sento understands human language and processes sentiment in real time – as fast as Twitter itself.
One of semantic analysis’ star players is called Named Entity Recognition. While not the sexiest term, what this scientific method proposes will definitely get you excited. Imagine having to sift through thousands of tweets in a desperate attempt to identify the most prominent people and places mentioned in this huge corpus of random, raw text related to the keyword “corruption”.
How about insider trading? sento will send you back names of organizations and businesses as well. Our Entity Extraction can also automatically detect events related to a keyword or a hashtag; say the recent London Tube Strike, helping you piece together a story. Discovery and serendipity abound.
Fans and foes, detractors and competitors are creating content every day about your brand or things that are important to your business. Concepts within these thousands of social media messages are being associated with your services and products: "battery life" "customer service" "location" "cancellation". Some are repeated by multiple users again and again.
Our concept extraction algorithm mines tweets related to your keywords to dig out the most relevant concept associated with your areas of interest, making spotting important issues effortless.
sento is the first text analytics tool to provide the ability to mine millions of tweets for what we call "Key Ideas" in real time. Key Ideas are the thoughts and impressions of users that are bubbling to the top of conversations en masse and are extremely useful for quickly understanding what the Twitter community is talking about when they talk about your topic of interest.
A quick look at our Key Ideas feature gives you a bird's eye view of the state of your brand, your competitor, your campaign, a popular news item -- or anything you'd like to track and learn more about.
What are people talking about when they talk about your brand? What themes emerge? Are there trends in their conversations? Being able to identify themes can help you quickly diffuse a bad trend of messages around your brand or help you ride the wave of particularly positive tweets that are gaining traction among users.
A good example of this would be searching for the keyword such as British Airways: users are talking about flight delays and sento’s semantic Topic Extraction module would quickly identify this in real time. BA would know the most important topics associated with its brand on Twitter updated by the second, allowing them to instantly respond to customer issues. The alternative offered by social media monitoring tools that don’t use semantic analysis? Sift through thousands of tweets or wait for a situation to bubble up – and its often too late to act. sento’s Topic Extraction is real time business intelligence for CRM, identifying issues and finding out what is more important to your audience.
This is going to sound confusing, but sentiment isn’t emotion… we promise. Sentiment Analysis finds the sentiment within a text but doesn’t necessarily tell you the emotion the user was feeling when he wrote the tweet. This becomes a problem in the case of sarcasm. I can write the hashtag #winning but the overall sentiment can be completely negative. Deep emotion mining helps you get past the text and into the feelings of the author.
At what time of day are users feeling the saddest? This may not be a good time to reach out to them. What are they feeling happy about? Get on their bandwagon and mix the things they are excited about into your campaign. Emotion mining is still an emerging science within semantic analysis and sento is one of the first tools to offer it as a standard feature.